水稲収量予察の作物学的研究 (予報) : (VI) 稔実歩合と収量予察との関係

抄録

Following significant relation between fruit bearing percentage and yield of rice was drawn out from the data, formerly reported on the experiments of pruning blades and roots in rice plant. (1) In this study, first of all, it was very necessary to establish an index by means of which fruited grains can be scientifically distinguished. As a result of researching the index from the stand-point of estimating rice yield, it was ascertained to be recommendable to use specific gravity of grains as the index and to take heavier grains than 1.06 of specific gravity as fruited ones. But this index holds true only in the case of dry grains (water content 12∼15%). (2) Correlation tables between fruit bearing percentage and yield, represented in persentage to the "Controls" respectively, are given in Fig. 1 and Fig. 2. Fruit bearing persentage can be seen to be very highly correlated with yield (r=0.99), except for the plots of No. 1∼8 in Fig. 1 and No. 1∼10 in Fig. 2. These two figures strongly suggest that in the period, at the latest, after the treatment date of No. 11 (Aug. 26, 5 days before heading), the yield of rice is mainly affected only by the fruit bearing percentage. The time of treatment on No. 11 is just the end of period when number of spikelets per ear increases, as seen in the former report (V), and there are no significant differences in number of spikelets per ears among all plots treated after it. Moreover, no significant differences are found in number of ears per plant among plots treated No.7 (Aug. 13, beginning of reproductive cell formation). Therefore; we can likely conclude that after the number of spikelets per eat and ears per plant are determined, fruit bearing percentage plays a leading role in determing the amount of rice-yield. (3) As the yield of rice is repesented by the product of its constitutional four factors, in the period after the number of ears per plant and spikelets per ear are determined it is controlled by two factors - fruit bearing percentage and weight of 1, 000 grains - which are not yet determined. In fact, however, the yield of rice, as mentioned above, is mainly affected only by fruit bearing percentage in that period. The reason why weight of 1, 000 grains has very little to do with constituting yield can be explained by its small variability. For example, Table 1 showing coefficients of variability of four factors indicates obviously that weight of 1, 000 grains is remarkably small in variability as compared with other three factors. And there are several further data supporting the stability of it. (4) From the results mentioned above, it has been made clear that one of the most important keys to solve the problem of yield-forecast in rice is to forecast the fruit bearing percentage as early as possible. Moreover, as entirely sterile grains can be recognized by the time shortly after heading, it seems to be the most urgent problem to forecast the percentage of lighter grains than 1.06 in specific gravity.